How the Doctor grades a slip

A Grade is a second opinion on a slip you already built, not a prediction. It starts from the exact saved ticket (legs, prices and stake), researches every leg in published sports records, and ends in one letter from A+ to F- with every answer and source on the report.

Records first: Jev, the research loop

Jev is the Doctor’s research loop. For each leg it identifies the game and the players, fetches published records (results, box scores, schedules, injury and lineup reports, venues and news), and answers each question only from what it fetched. Historical results are calculated from the records where the records support them. An answer that no record backs is not written.

Each answer keeps its source and a support reading: how strongly that record backs the pick. Conflicting records are kept as a conflict, not averaged away. A language model writes the short plain-English read at the end, but it may only use the facts already in the report, cites them paragraph by paragraph, and does not set the letter.

100 questions in 10 groups

Every leg is checked against the same bank of 100 sports-record questions, ten in each group. The depth changes how hard Jev looks, never the questions.

  1. 01 Participation & Availability
  2. 02 Lineups, Roles & Opportunities
  3. 03 Recent Competitive Results
  4. 04 Exact-Statistic History
  5. 05 Opposition & Field Profile
  6. 06 Direct Matchup History
  7. 07 Schedule, Rest & Workload
  8. 08 Venue, Surface & Conditions
  9. 09 Competition Format & Position
  10. 10 Periods, Sequences & Live State

“X of Y applicable”

Questions that cannot apply are removed before anything is counted: 97 apply to football, 95 to tennis and 98 to basketball, hockey, baseball, soccer, MMA and boxing. Before a game, a lineup question with no published lineup counts as not applicable. The report then says how many of the applicable questions were answered, for example “79 of 97 applicable”.

  • Answered

    A record answered it. The answer carries its source and how strongly it supports the pick.

  • Unavailable

    The question applies, but no record could be checked in time. It is never filled in with a guess.

  • Not applicable

    The question cannot apply to this sport or this slip, for example a lineup question before the lineup is published. It leaves the count.

The score and the evidence score

The answers and the prices become four factors and one score out of 100.

  • Value

    30 points

    Your saved price against the market consensus from other sportsbooks.

  • Research

    40 points

    The evidence score: 70% coverage (how many applicable questions were answered) and 30% support (how strongly the answered records back the picks).

  • Slip construction

    20 points

    How the legs fit together: leg count, legs that depend on each other, and legs that cannot all win.

  • Stake fit

    10 points

    The stake against your own saved unit.

The letter

The score maps to a letter on a fixed table. The cut points come from the spread of real production Grades: about the top 10% earn an A and the bottom 10% an F.

A+ 89+ · A 88 · A- 83-87 · B+ 81-82 · B 80 · B- 78-79 · C+ 76-77 · C 75 · C- 73-74 · D+ 72 · D 71 · D- 68-70 · F+ 67 · F 65-66 · F- 64 and below

No A without a price at or better than the market consensus and a Research score of 70 or more; without both, the score stops at 82, the top of B+. Each Grade also shows where it ranks, for example “Scores higher than 59% of slips graded”.

Depths

Deeper depths run more research rounds, read more pages and look further back in the news. Each depth has a hard time limit.

  • Fast

    Every account, including Free

    One research round over the records for each leg. No browser follow-up.

    1 round · up to 5 pages · 3-day news window · limit 150 to 210 seconds

  • Smart

    Starter and up

    A second round, then a browser follow-up on public sports and news sites for answers the records left missing.

    2 rounds · up to 18 pages · 7-day news window · limit 5 minutes

  • Deep

    Pro and up

    A third round. The browser follow-up also chases conflicting reports, uncertain playing time and matchup detail.

    3 rounds · up to 40 pages · 14-day news window · limit 10 minutes

  • Elite

    Elite

    A fourth round, then an independent counter-evidence pass challenges the research result.

    4 rounds · up to 80 pages · 30-day news window · limit 15 minutes

What a Grade can’t know

Injuries break late. Lines move after the prices were read. Records describe the past, and the next game can go the other way. If the research cannot finish, the Grade says “Inconclusive” instead of guessing, and every unanswered question is shown as unavailable. A Grade is a read on price, evidence and build, not a win probability, and never a guarantee.

For the full responsible-use policy, see responsible use & terms.

Questions people ask

How does BetSlip Doctor grade a bet slip?
Jev, the Doctor’s research loop, checks every leg against a bank of 100 sports-record questions in 10 groups, answering only from records it fetched. The answers and your price against the market become a score out of 100 (Value 30, Research 40, Slip construction 20, Stake fit 10), and the score maps to a letter from A+ to F-.
What does “79 of 97 applicable” mean?
Questions that cannot apply to the sport or the slip are removed first (97 apply to football, 95 to tennis, 98 to most other sports). The first number is how many of the rest Jev answered from records.
What is the evidence score?
The Research factor (40 points): 70% is coverage, how many applicable questions were answered, and 30% is support, how strongly the answered records back the picks.
Is a Grade a win probability?
No. A Grade reads price, evidence and build. It never predicts a result and never guarantees one.
What changes with a deeper depth?
The 100 questions stay the same. Deeper depths run more research rounds, read more pages and look further back in the news: Fast 1 round, Smart 2, Deep 3, Elite 4 plus a counter-evidence pass.